Maticlib

A Python automation library for creating intelligent AI agents

MIT License Downloads PyPI version

Features

🤖

Core Agent Framework

MaticGraph workflow engine with stateful & stateless execution, conditional routing, and multi-state support

🔄

LLM Integration

Support for Google Gemini, Mistral AI with unified response models and multi-turn conversations

🛠️

Developer Experience

Built-in error handling, type-safe design, and environment variable support for seamless integration

📊

Response Handling

Standardized response models with token usage tracking and modality-specific metrics

🎯

Advanced Capabilities

Conditional branching, parallel node execution, checkpoint system, and workflow templates

High Performance

Async support, minimal dependencies, and optimized for production workloads

Quick Example

from maticlib.llm.google_genai import GoogleGenAIClient

# Initialize client
client = GoogleGenAIClient(api_key="YOUR_API_KEY")

# Make a request
response = client.complete("Hello! Tell me about Python")
print(response.content)